canqc: Quebec Provincial Highways

[rickmastfan67]

Mapcat, pick a place you want the boarder to be, post it here, and I'll update my copy of ON-63 with it after making tweaks for the island addition.

I suggest http://www.openstreetmap.org/?lat=46.711643&lon=-79.100525

Fine with me.

Think this fix is worthy of a 'news update', or can it go under the radar since no new 'visible' points are being added to the island?

[oscar]

^ I think the change is minor enough that it won't require an Updates entry.

I'll pull in the updated QC 101 file, with the new border point, along with TCH and canqc changes from the first round of canqc peer review, and cansk changes from Saskatoon. I'll probably do it tomorrow.

[rickmastfan67]

^ I think the change is minor enough that it won't require an Updates entry.

I'll pull in the updated QC 101 file, with the new border point, along with TCH and canqc changes from the first round of canqc peer review, and cansk changes from Saskatoon. I'll probably do it tomorrow.

Just pushed the ON fix.
https://github.com/TravelMapping/HighwayData/pull/3116

[oscar]

https://github.com/TravelMapping/HighwayData/pull/3122 for the QC changes, including to qc.tchkir.

[michih]

Also some LABEL_SELFREF errors but I think that they just need to be marked FP: http://travelmapping.net/devel/datacheck.php?rg=QC

[oscar]

Also some LABEL_SELFREF errors but I think that they just need to be marked FP: http://travelmapping.net/devel/datacheck.php?rg=QC

I think those are "lollipop" routes that intersect themselves. When I'm back home from New Jersey, I'll mark them as FPs.

[si404]

Close enough. As I've said in other contexts, while I can be (overly?) perfectionist on placement of labeled waypoints, I don't view shaping points as a high-precision exercise, especially when they're shared with two other route files.

Noted - I won't push shaping points.

I would leave QC 104 as is:

-- I've always felt that an exception to "one point per interchange" is appropriate where two TM-mapped routes intersect with a third at different places within an interchange, so we have clearer graph connections

I agree. I put the question mark at the end of the the comment as I would personally map it as it is, even if the rules would probably prefer one point.

Or better still, ChLacSM.

Indeed!

Anyway, back to the review (there seems to be some 'OSM drift' here):

QC111:
 - ChRou is off where OSM puts the intersection
 - ChBec is slightly off where OSM puts the intersection
 - Rang1 -> Rang1e
 - Rang5e is very slightly off where OSM puts the intersection
 - RteLit -> ChCol?
 - ChLem is very slightly off where OSM puts the intersection
 - ChJosLan -> ChBoi (also very slightly off)
 - QC399 is slightly off where OSM puts the intersection
 - ChLacTru is very slightly off where OSM puts the intersection
 - RueChi is off where OSM puts the intersection
 - ChBaieCan is off where OSM puts the intersection
 - QC390 is very slightly off where OSM puts the intersection
 - CheNord -> ChNord
 - ChPio is slightly off where OSM puts the intersection (also check name)
 - ChPri is off where OSM puts the intersection
 - QC393_S is very slightly off where OSM puts the intersection
 - QC393_N is slightly off where OSM puts the intersection
 - Rte45 -> RteDupCle?
 - RueCom is very slightly off where OSM puts the intersection
 - ChPar_W is slightly off where OSM puts the intersection

QC112:
 - QC138 is slightly off where OSM puts the intersection
 - A-20/132 -> A-20?
 - separate GSJ with Rue Riverside just east of the Autoroute
 - exit numbers: aren't they QC116's?
 - QC223_N is slightly off where OSM puts the intersection
 - QC133 is slightly off where OSM puts the intersection
 - QC233 is slightly off where OSM puts the intersection
 - RangMon is slightly off where OSM puts the intersection
 - RueOst is slightly off where OSM puts the intersection
 - ChePic is very slightly off where OSM puts the intersection
 - QC241_S is very slightly off where OSM puts the intersection
 - QC243_S is slightly off where OSM puts the intersection
 - QC245 is very slightly off where OSM puts the intersection
 - CheOrfLac is slightly off where OSM puts the intersection
 - CheRoy is slightly off where OSM puts the intersection
 - A-55(33) is slightly off where OSM puts the intersection
 - A-55(33) -> A-55 ?
 - A-10(123) is slightly off where OSM puts the intersection
 - A-10(123A) -> QC249 (not a junction on A-10 or A-55, AFAICS)
 - A-10(128) is slightly off where OSM puts the intersection
 - A-410 is very slightly off where OSM puts the intersection
 - move A-610 to the centre of the roundabout
 - move RueStJean from the overpass to the link road
 - ChBas is slightly off where OSM puts the intersection
 - QC255 is slightly off where OSM puts the intersection
 - ChSta is very slightly off where OSM puts the intersection
 - 4eRang is slightly off where OSM puts the intersection
 - QC257_S is slightly off where OSM puts the intersection
 - QC257_N is very slightly off where OSM puts the intersection
 - ChPare is slightly off where OSM puts the intersection
 - QC263_N is very slightly off where OSM puts the intersection
 - QC263_S is off where OSM puts the intersection
 - *OldQC112_W - would it make more sense to have as non-starred point at Ancienne Route 112?
 - *OldQC112_W is very slightly off where OSM has the highway
 - *OldQC112_E - would it make more sense to have as non-starred point at Boulevard des Mineurs?
 - QC165 is very slightly off where OSM puts the intersection
 - QC269_S is very slightly off where OSM puts the intersection
 - QC271_N is off where OSM puts the intersection
 - QC271_S is slightly off where OSM puts the intersection
 - AvStJos is very slightly off where OSM puts the intersection
 - QC276 is very slightly off where OSM puts the intersection
 - RangBasStA is slightly off where OSM puts the intersection
 - RangBasStA -> RangBSA?
 - QC173_S is very slightly off where OSM puts the intersection
 - QC173_N is slightly off where OSM puts the intersection
 - A-73 is slightly off where OSM puts the intersection
 - AvPri is very slightly off where OSM puts the intersection
 - QC275 is slightly off where OSM puts the intersection

QC113:
 - most points are very slightly off where OSM puts the intersection - only going to explicitly point out the most egregious
 - QC386 is off where OSM puts the intersection
 - 3eRue is off where OSM puts the intersection
 - 3eRue -> 2eRue
 - BoulQue is off where OSM puts the intersection
 - ChLacQue is off where OSM puts the highway (minor road not on it)
 - ChLacCam is off where OSM puts the intersection
 - ChRivOSul is off where OSM puts the highway (minor road not on it)
 - ChRivOSul -> ChROS?
 - RuePeu is off where OSM puts the highway (minor road not on it)
 - PosAbi is off where OSM puts the highway (minor road not on it)
 - CFL209N is marked as a bridge (albeit for miles) on OSM - does it actually intersect
 - CFL209N is called Route Forestiere R1009 on OSM - relabel RF1009?
 - Parc, ChLacQue, ChRivOSul, RuePeu and PosAbi don't have intersections on OSM. Not saying there isn't one here, but that they seem to be minor.

[yakra]

I agree. I put the question mark at the end of the the comment as I would personally map it as it is, even if the rules would probably prefer one point.

Would also map as-is.

[oscar]

Just one comment on a tricky issue, until I have some time to address the others.

...

QC112:

...

 - A-10(123A) -> QC249 (not a junction on A-10 or A-55, AFAICS)

It's an indirect junction with the Autoroutes, since Autoroute traffic heading toward QC 249 is directed onto QC 112 (which runs on a frontage road on both sides of A-10/55) to exit onto QC 249, and traffic from QC 249 briefly uses QC 112 to get to the Autoroutes. The A-10 and A-55 route files each have a 123A point for the QC 249 intersection. Kind of how we handle exits from express lanes onto collector/distributor roads. except the collector/distributor for A-10/55 is QC 112.

[si404]

Can we get another opinion on this?

[Markkos1992]

Just one comment on a tricky issue, until I have some time to address the others.

...

QC112:

...

 - A-10(123A) -> QC249 (not a junction on A-10 or A-55, AFAICS)

It's an indirect junction with the Autoroutes, since Autoroute traffic heading toward QC 249 is directed onto QC 112 (which runs on a frontage road on both sides of A-10/55) to exit onto QC 249, and traffic from QC 249 briefly uses QC 112 to get to the Autoroutes. The A-10 and A-55 route files each have a 123A point for the QC 249 intersection. Kind of how we handle exits from express lanes onto collector/distributor roads. except the collector/distributor for A-10/55 is QC 112.

I am thinking that QC 112 should have shaping points added between the junctions on A-10 and A-55 without changing the visible points.  This would treat it a bit like FL 84, which mostly runs as a Frontage Road along I-75 and I-595, but I would definitely not consider it as concurrent with (hence continual breaking of it).

I'd recommend that rickmastfan67 look into my thoughts here.

[oscar]

I am thinking that QC 112 should have shaping points added between the junctions on A-10 and A-55 without changing the visible points.  This would treat it a bit like FL 84, which mostly runs as a Frontage Road along I-75 and I-595, but I would definitely not consider it as concurrent with (hence continual breaking of it).

I already added the shaping points (one on each side of the QC 249 overpass, by about 0.2 mi.) when I synched up the A-10/A-55/QC 112/QC 249 waypoints. My own map shows that I traveled A-10 and A-55 in that area, but was not credited with mileage (km-age?) on the parallel segment of QC 112.

[yakra]

Just one comment on a tricky issue, until I have some time to address the others.

...

QC112:

...

 - A-10(123A) -> QC249 (not a junction on A-10 or A-55, AFAICS)

It's an indirect junction with the Autoroutes, since Autoroute traffic heading toward QC 249 is directed onto QC 112 (which runs on a frontage road on both sides of A-10/55) to exit onto QC 249, and traffic from QC 249 briefly uses QC 112 to get to the Autoroutes. The A-10 and A-55 route files each have a 123A point for the QC 249 intersection. Kind of how we handle exits from express lanes onto collector/distributor roads. except the collector/distributor for A-10/55 is QC 112.

Reminds me of Texas, though not as straightforward as some of the examples I've dealt with. In the A10/A55 context, I'd call that a quarter interchange: worthy of a waypoint, but just barely.
In the QC112 context, I'd probably name it after the locally intersecting route, QC249, rather than that one slip ramp farther away.

[froggie]

My thoughts, since I've traveled both routes a fair bit (live less than an hour-and-a-half away):

- Rename the point on QC112 to QC249 as Si originally suggests.
- Move the QC112/QC249 point to the southern junction (i.e. on QC112's eastbound lanes)
- If the QC112/QC249 point is moved to the eastbound junction, a second point (for the westbound ramps) may or may not be warranted.  I'm indifferent either way.
- Use shaping points on QC112 as needed to tweak it to follow the eastbound lanes.
- Move the 123 point on A-10 and A-55 closer to where the exit ramp is.  While the north/eastbound exit ramp can technically be used to access 112 and (via a downstream U-turn) 249, its primary purpose is to send trucks to the weigh station.

[oscar]

- Rename the point on QC112 to QC249 as Si originally suggests.

Will do.

- Move the QC112/QC249 point to the southern junction (i.e. on QC112's eastbound lanes)
- If the QC112/QC249 point is moved to the eastbound junction, a second point (for the westbound ramps) may or may not be warranted.  I'm indifferent either way.
- Use shaping points on QC112 as needed to tweak it to follow the eastbound lanes.

Why the eastbound, rather than the westbound, lanes? And why not follow our usual practice for one-way couplets to put waypoints about halfway between the roadways, which in this case would be in the A-10/55 median? Moving QC112's points to the eastbound lanes would also mean no graph connection between A-10/55, QC 112, and QC 249. And the HB also would not show QC249 as an intersecting route for A-10/55, as I think it should considering QC 249 signage for the slip ramps between A-10/55 and QC 112.

- Move the 123 point on A-10 and A-55 closer to where the exit ramp is.  While the north/eastbound exit ramp can technically be used to access 112 and (via a downstream U-turn) 249, its primary purpose is to send trucks to the weigh station.

I think if we move the 123 point at all (I'm trying to minimize changes to active TCH and A- routes, lest we take forever to make everything perfect), it should be in the middle of the A-10/A-55/QC 112 interchange, rather than a ramp location. Something like this, treating that interchange kinda like a trumpet interchange.